US7119816B2 - System and method for whiteboard scanning to obtain a high resolution image - Google Patents

Abstract

This invention is directed toward a system and method for scanning a scene or object such as a whiteboard, paper document or similar item. More specifically, the invention is directed toward a system and method for obtaining a high-resolution image of a whiteboard or other object with a low-resolution camera. The system and method of the invention captures either a set of snapshots with overlap or a continuous video sequence, and then stitches them automatically into a single high-resolution image. The stitched image can finally be exported to other image processing systems and methods for further enhancement.

Description

BACKGROUND

1. Technical Field

This invention is directed toward a system and method for obtaining a high-resolution image of a whiteboard or other object. More specifically, this invention is directed toward a system and method for obtaining a high-resolution image of a whiteboard or similar object with a low-resolution camera.

2. Background Art

The many advances in technology have revolutionalized the way meetings are conducted. For instance, many knowledge workers attend meetings with their notebook computers. Additionally, many meetings nowadays are distributed—that is, the meeting participants are not physically co-located and meet via video-conferencing equipment or via a network with images of the meeting participants taken by a web camera (web cam) and transferred over the network.

One fairly common device used in meetings is the conventional whiteboard that is written on during a meeting by the meeting participants. Alternately, an easel with a white paper pad is also used. Many meeting scenarios use a whiteboard extensively for brainstorming sessions, lectures, project planning meetings, patent disclosures, and so on. Note-taking and copying what is written on the board or paper often interferes with many participants' active contribution and involvement during these meetings. As a result, efforts have been undertaken to capture this written content in some automated fashion. One such method is via capturing an image of the written content. There are, however, issues with this approach to capturing the content of a whiteboard or paper document.

Although a typical lap top computer is sometimes equipped with a built-in camera, it is normally not possible to copy images of annotations of a fruitful brainstorming session on a whiteboard because the typical built-in laptop camera has a maximum resolution 640×480 pixels that is not high enough to produce a readable image of the whiteboard.

Likewise, in the distributed meeting scenario where a meeting participant has a document only in paper form to share with other remote meeting participants, a web cam, which typically has a maximum resolution of 640×480 pixels, is unable to produce a readable image of the paper document to provide to the other participants.

Hence, the current technology is lacking in capturing whiteboard or other document data for the above-mentioned scenarios, and many other similar types of situations.

SUMMARY

The invention is directed toward a system and method that produces a high-resolution image of a whiteboard, paper document or similar planar object with a low-resolution camera by scanning the object to obtain multiple images and then stitching these multiple images together. By zooming in (or approaching to the whiteboard physically) and taking smaller portions of the object in question at a given resolution, a higher resolution image of the object can be obtained when the lower-resolution images are stitched together.

The planar object image enhancing system and method for creating a high-resolution image from low-resolution images can run in two modes: snapshot or continuous. Although the image acquisition procedure differs for the two operation modes, the stitching process is essentially the same.

In snapshot mode, one starts by acquiring a snapshot from the upper left corner of the object such as a whiteboard, a second by pointing to the right but having overlap with previous snapshot, and so on until reaching the upper right corner; moving the camera lower and taking a snapshot, then taking another one by pointing to the left, and so on until reaching the left edge. The process continues in this horizontally flipped S-shaped pattern until the lower border is captured. Successive snapshots must have overlap to allow later stitching, and this is assisted by providing visual feedback during acquisition.

In continuous mode, the user takes images also starting from the upper left corner but in this case continuously following the same S-shaped pattern discussed above without stopping to capture an image. The difference from the snapshot mode is that the user does not need to wait and position the camera anymore before taking a snapshot. The continuous image acquisition also guarantees overlap between successive images assuming a sufficient capture rate. However, motion blur may cause the final stitched image look not as crisp as those obtained with snapshot mode. In order to reduce the blur, the camera exposure time should be set to a small value.

Of course, other acquisition patterns besides the above-mentioned ones can also be used. For example, one can start from the upper left corner, from the lower left corner, or from the lower right corner of the whiteboard or other planar object when capturing the overlapping images.

The mathematic foundation behind the invention is that two images of a planar object, regardless the angle and position of the camera, are related by a plane perspectivity, represented by a 3×3 matrix called homography H. The homography defines the relationship between the points of one image and points in a subsequent image. This relationship is later used to stitch the images together into a larger scene. It typically is a simple linear projective transformation. At least 4 pairs of point matches are needed in order to determine homography H.

Given this, the stitching process involves first, for each image acquired, extracting points of interest. In one embodiment of the image enhancement system and method of the invention, a Plessey corner detector, a well-known technique in computer vision, is used to extract these points of interest. It locates corners corresponding to high curvature points in the intensity surface if one views an image as a 3D surface with the third dimension being the intensity. However, other conventional methods of detecting the points of interest could also be used. These include, for example, a Moravec interest detector.

Next, an attempt is made to match the extracted points with those from a previous image. For each point in the previous image, a 15×15 pixel window is chosen (although another sized window could be chosen) centered on the point under consideration, and the window is compared with windows of the same size, centered on the points in the current image. A zero-mean normalized cross correlation between two windows is computed. If the intensity values of the pixels in each window are rearranged as a vector, the correlation score is equivalent to the cosine angle between the two intensity vectors. The correlation score ranges from −1, for two windows that are not similar at all, to 1, for two windows which are identical. If the largest correlation score exceeds a prefixed threshold (0.707 in one working embodiment of the invention), then the associated point in the current image is considered to be the match candidate to the point in the previous image under consideration. The match candidate is retained as a match if and only if its match candidate in the previous image happens to be the point being considered. This symmetric test reduces many potential matching errors.

The set of matches established by correlation usually contains false matches because correlation is only a heuristic and only uses local information. Inaccurate location of extracted points because of intensity variation or lack of strong texture features is another source of error. The geometric constraint between two images is the homography constraint. If two points are correctly matched, they must satisfy this constraint, which is unknown in this case. If the homography between the two images is estimated based on a least-squares criterion, the result could be completely wrong even if there is only one false match. This is because least-squares is not robust to outliers (erroneous data). A technique based on a robust estimation technique known as the least median squares was developed to detect both false matches and poorly located corners, and simultaneously estimate the homography matrix H.

The aforementioned optimization is performed by searching through a random sampling in the parameter space to find the parameters yielding the smallest value for the median of squared residuals computed for the entire data set. From the smallest median residual, one can compute a so-called robust standard deviation {circumflex over (σ)}, and any point match yielding a residual larger than, say, 2.5{circumflex over (σ)} is considered to be an outlier and is discarded. Consequently, it is able to detect false matches as many as 49.9% of the whole set of matches.

This incremental matching procedure of the stitching process stops when all images have been processed.

Because of the incremental nature, cumulative errors are unavoidable. For higher accuracy, one needs to adjust H's through global optimization by considering all the images simultaneously. Take an example of a point that is matched across three views. In the incremental case, they are considered as two independent pairs, the same way as if they were projections of two distinct points in space. In the global optimization, the three image points are treated exactly as the projections of a single point in space, thus providing a stronger constraint in estimating the homographies. Therefore, the estimated homographies are more accurate and more consistent.

Once the geometric relationship between images (in terms of homography matrices H's) are determined, all of the images can be stitched together as a single high-resolution image. There are several options, and in one working embodiment of the invention a very simple one was implemented. In this embodiment, the first image is used as the reference frame of the final high-resolution image, and original images are successively matched to the reference frame. If a pixel in the reference frame appears several times in the original images, then the one in the newest image is retained.

The image enhancing system and method according to the invention has many advantages. For instance, the invention can produce a high-resolution image from a low-resolution set of images. Hence, only a low-resolution camera is necessary to create such a high-resolution image. This results in substantial cost savings. Furthermore, high-resolution images can be obtained with typical equipment available and used in a meeting. No specialized equipment is necessary.

In addition to the just described benefits, other advantages of the present invention will become apparent from the detailed description which follows hereinafter when taken in conjunction with the drawing figures which accompany it.

DESCRIPTION OF THE DRAWINGS

The file of this patent contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the U.S. Patent and Trademark Office upon request and payment of the necessary fee.

The specific features, aspects, and advantages of the invention will become better understood with regard to the following description, appended claims, and accompanying drawings where:

FIG. 1 is a diagram depicting a general purpose computing device constituting an exemplary system for implementing the invention.

FIG. 2 is a general flow diagram of the planar object image enhancing system and method for creating a high-resolution image from low-resolution images.

FIG. 3 is an exemplary user interface employed in one working embodiment of the system and method according to the invention.

FIG. 4A is an illustration of the snapshot image acquisition mode of the image enhancing system and method according to the invention.

FIG. 4B is an illustration of the continuous image acquisition mode of the image enhancing system and method according to the invention.

FIG. 5 is a flow diagram depicting the process of extracting points of interest in the image enhancing system and method according to the invention.

FIG. 6 is an image showing an example of extracted points of interest, indicated by a + of the image enhancing system and method according to the invention.

FIG. 7 is a flow diagram depicting the process of matching the points of interest between images in the image enhancing system and method according to the invention.

FIG. 8 is a flow diagram depicting the process of estimating the homography between two images in the image enhancing system and method according to the invention.

FIG. 9 is a flow diagram depicting the process of stitching together images to obtain a high-resolution image in the image enhancing system and method according to the invention.

FIG. 10 is a flow diagram depicting an alternate process of matching points of interest in the system and method according to the present invention.

FIG. 11 is a series of images of portions of a whiteboard in an office.

FIG. 12 is a stitched image from the series of images shown inFIG. 11 determined by the image enhancing system and method according to the invention. The cumulative error is visible at the left border.

FIG. 13 shows three images of a paper document.

FIGS. 14A and 14B show a comparison of the stitched image from those shown ifFIG. 13 provided by one working embodiment of the invention (FIG. 14A) and the same document captured by the same camera as a single image (FIG. 14B).

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In the following description of the preferred embodiments of the present invention, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. It is understood that other embodiments may be utilized and structural changes may be made without departing from the scope of the present invention.

1.0 Exemplary Operating Environment

FIG. 1 illustrates an example of a suitablecomputing system environment100 on which the invention may be implemented. Thecomputing system environment100 is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of the invention. Neither should thecomputing environment100 be interpreted as having any dependency or requirement relating to any one or combination of components illustrated in theexemplary operating environment100.

The invention is operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well known computing systems, environments, and/or configurations that may be suitable for use with the invention include, but are not limited to, personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

With reference toFIG. 1, an exemplary system for implementing the invention includes a general purpose computing device in the form of acomputer110. Components ofcomputer110 may include, but are not limited to, aprocessing unit120, asystem memory130, and asystem bus121 that couples various system components including the system memory to theprocessing unit120. Thesystem bus121 may be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus also known as Mezzanine bus.

Computer110 typically includes a variety of computer readable media. Computer readable media can be any available media that can be accessed bycomputer110 and includes both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer readable media may comprise computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed bycomputer110. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of the any of the above should also be included within the scope of computer readable media.

Thesystem memory130 includes computer storage media in the form of volatile and/or nonvolatile memory such as read only memory (ROM)131 and random access memory (RAM)132. A basic input/output system133 (BIOS), containing the basic routines that help to transfer information between elements withincomputer110, such as during start-up, is typically stored inROM131.RAM132 typically contains data and/or program modules that are immediately accessible to and/or presently being operated on by processingunit120. By way of example, and not limitation,FIG. 1 illustratesoperating system134, application programs135,other program modules136, andprogram data137.

Thecomputer110 may also include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only,FIG. 1 illustrates ahard disk drive141 that reads from or writes to non-removable, nonvolatile magnetic media, amagnetic disk drive151 that reads from or writes to a removable, nonvolatilemagnetic disk152, and anoptical disk drive155 that reads from or writes to a removable, nonvolatileoptical disk156 such as a CD ROM or other optical media. Other removable/non-removable, volatile/nonvolatile computer storage media that can be used in the exemplary operating environment include, but are not limited to, magnetic tape cassettes, flash memory cards, digital versatile disks, digital video tape, solid state RAM, solid state ROM, and the like. Thehard disk drive141 is typically connected to thesystem bus121 through anon-removable memory interface such asinterface140, andmagnetic disk drive151 andoptical disk drive155 are typically connected to thesystem bus121 by a removable memory interface, such asinterface150.

The drives and their associated computer storage media discussed above and illustrated inFIG. 1, provide storage of computer readable instructions, data structures, program modules and other data for thecomputer110. InFIG. 1, for example,hard disk drive141 is illustrated as storingoperating system144,application programs145,other program modules146, andprogram data147. Note that these components can either be the same as or different fromoperating system134, application programs135,other program modules136, andprogram.data137.Operating system144,application programs145,other program modules146, andprogram data147 are given different numbers here to illustrate that, at a minimum, they are different copies. A user may enter commands and information into thecomputer110 through input devices such as akeyboard162 andpointing device161, commonly referred to as a mouse, trackball or touch pad. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner, or the like. These and other input devices are often connected to theprocessing unit120 through auser input interface160 that is coupled to thesystem bus121, but may be connected by other interface and bus structures, such as a parallel port, game port or a universal serial bus (USB). Amonitor191 or other type of display device is also connected to thesystem bus121 via an interface, such as avideo interface190. In addition to the monitor, computers may also include other peripheral output devices such asspeakers197 andprinter196, which may be connected through an outputperipheral interface195. Of particular significance to the present invention, a camera192 (such as a digital/electronic still or video camera, or film/photographic scanner) capable of capturing a sequence ofimages193, can also be included as an input device to thepersonal computer110. Further, while just one camera is depicted, multiple cameras could be included as input devices to thepersonal computer110. Theimages193 from the one or more cameras are input into thecomputer110 via anappropriate camera interface194. Thisinterface194 is connected to thesystem bus121, thereby allowing the images to be routed to and stored in theRAM132, or one of the other data storage devices associated with thecomputer110. However, it is noted that image data can be input into thecomputer110 from any of the aforementioned computer-readable media as well, without requiring the use of thecamera192.

Thecomputer110 may operate in a networked environment using logical connections to one or more remote computers, such as aremote computer180. Theremote computer180 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to thecomputer110, although only amemory storage device181 has been illustrated inFIG. 1. The logical connections depicted inFIG. 1 include a local area network (LAN)171 and a wide area network (WAN)173, but may also include other networks. Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets and the Internet.

When used in a LAN networking environment, thecomputer110 is connected to theLAN171 through a network interface oradapter170. When used in a WAN networking environment, thecomputer110 typically includes amodem172 or other means for establishing communications over theWAN173, such as the Internet. Themodem172, which may be internal or external, may be connected to thesystem bus121 via theuser input interface160, or other appropriate mechanism. In a networked environment, program modules depicted relative to thecomputer110, or portions thereof, may be stored in the remote memory storage device. By way of example, and not limitation,FIG. 1 illustrates remote application programs185 as residing onmemory device181. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.

The exemplary operating environment having now been discussed, the remaining parts of this description section will be devoted to a description of the program modules embodying the invention.

2.0 System and Method for Whiteboard Scanning to obtain a High Resolution Image

The invention is directed toward a system and method of converting the content of a regular whiteboard, paper document or similar planar object into a single high-resolution image that is composed of stitched together lower-resolution images.

2.1 General Overview

A general flow chart of the system and method according to the invention is shown inFIG. 2. The system begins by acquiring images of portions of a whiteboard, paper document or other subject of interest in a prescribed overlapping pattern, with a still or video camera, as shown inprocess action202. The acquired images are represented in a digital form, consisting of an array of pixels. Once an image has been obtained, the points of interest are extracted from this image (process action204). The points of interest extracted inprocess action204 are matched with the previous image of the subject, as shown inprocess action206. Outlying points of interest are rejected, and a homography H is estimated (process action208). This continues for each image in turn until the final image captured is reached (process action210), at which time the homographies may be adjusted through global optimization (process action212). The acquired images are then stitched together (process action214) to create a high-resolution composite image of the item of interest.

The general system and method according to the invention having been described, the next paragraphs provide details of the aforementioned process actions.

2.2 Image Acquisition

The system can acquire images in two modes: snapshot or continuous. Although the image acquisition procedure differs for the two operation modes, the stitching process is essentially the same. In snapshot mode, one starts by taking a snapshot from the upper left corner, a second by pointing to the right but having overlap with previous snapshot, and so on until reaching the upper right corner; moving the camera lower and taking a snapshot, then taking another one by pointing to the left, and so on until reaching the left edge. The process continues in this horizontally flipped S-shaped pattern until the lower border is captured. Successive snapshots must have overlap to allow later stitching, and this is assisted by providing visual feedback during acquisition. In one working embodiment of the system and method according to the invention, an image overlap of approximately 50% is suggested, but the system works with much less overlap in sacrificing the accuracy of the stitching quality, e.g., 5 to 20%.FIG. 3 depicts auser interface302 of one working embodiment of the invention. In theviewing region304, both the previously acquired image and the current image or video is displayed. In order to facilitate the image acquisition, approximately half of the previously acquiredimage308 is shown as opaque, while the other half, which is in theoverlapping region310, is shown as semi-transparent. The current live video is also shown as half opaque and half semi-transparent. This guides the user to take successive images with overlap. Note that the alignment does not need to be precise. The system and method according to the invention will align them. There are also afew buttons306a,306b,306c,306dto indicate the direction in which the user wants to move the camera (down, up, left, right). The overlapping regions change depending on the direction. In one embodiment of the invention, the default behavior was designed such that only the “down” button is necessary to realize image acquisition in the desired patternFIG. 4A illustrates this image acquisition process.

In continuous mode, the user takes images also starting from the upper left corner but in this case continuously following the S-shaped pattern without stopping to capture a specific image as illustrated inFIG. 4B. The difference from the snapshot mode is that the user does not need to wait and position the camera before taking a snapshot. The continuous image acquisition also guarantees overlap between successive images. However, motion blur may cause the final stitched image look not as crisp as those obtained with snapshot mode. In order to reduce the blur, the camera exposure time should be set to a small value.

It should be noted that the pattern of acquisition for either the snap shot or continuous mode could be performed in other prescribed patterns as long as there is an overlap between successive images. For example, the pattern could start at the right upper corner and move to the left and downward. Or, similarly, the pattern could start at the lower right corner and move to the left and upwards.

The stitching process works very much in a similar way in both image acquisition operation modes, and is illustrated inFIG. 2.

2.3 Extracting Points of Interest

Referring toFIGS. 2 and 5, for each image acquired, points of interest are extracted. A Plessey corner detector, which is a well-known technique in computer vision, is used in one embodiment of the image enhancement system and method of the invention. As shown inFIG. 5,process action502, an image is input. The Plessey corner detector locates corners corresponding to high curvature points in the intensity surface if one views an image as a 3D surface with the third dimension being the intensity (process action504). As shown inprocess action506, these points corresponding to the corners are extracted as the points of interest. An example is shown inFIG. 6, where the extracted points are displayed in red +.

2.4 Matching Points of Interest

Next, as shown inFIG. 7, an attempt is made to match the extracted points with those from the previous image as is shown inprocess action702. For each point in the previous image, a 15×15 pixel window is chosen (although a different sized window could be chosen) centered on the point under consideration, and the window is compared with windows of the same size, centered on the points in the current image (processactions704 through708). A zero-mean normalized cross correlation between two windows is computed in order to make this comparison. If the intensity values of the pixels in each window are rearranged as a vector, the correlation score is equivalent to the cosine angle between the two intensity vectors. The correlation score ranges from −1, for two windows that are not similar at all, to 1, for two windows that are identical. As shown inprocess action710, if the largest correlation score found for each point in the previous image exceeds a prefixed threshold (0.707 in one working embodiment of the invention), then the associated point in the current image is considered to be the match candidate to the point in the previous image under consideration. The match candidate is retained as a match if and only if its match candidate in the previous image happens to be the point being considered. That is, by reversing the role of the two images, an attempt is made to find the best match in the previous image for the match candidate in the current image; if the best match in the previous image is the point under consideration, this point and the match candidate are considered to be matched; otherwise, the match candidate is discarded, and there is no match for the point under consideration. This symmetric test reduces many potential matching errors.

2.5 Rejecting Outliers and Estimating the Homography.

The mathematic foundation behind the invention is that two images of a planar object, regardless of the angle and position of the camera, are related by a plane perspectivity, represented by a 3×3 matrix called homography H. The homography defines the relationship between the points of one image and points in another image. This relationship is later used to stitch the images together into a larger scene. More precisely, let m₁=[u₁, v₁]^Tand m₂=[u₂, v₂]^Tbe a pair of corresponding points, and use the notation ˜ for {tilde over (m)}=[u, v,1]^T, then
{tilde over (m)}₂=λH{tilde over (m)}₁  (1)
where λ is a scalar factor. That is, H is defined up to a scalar factor. At least 4 pairs of point matches are needed in order to determine a homography H between two images.

The set of matches established by correlation usually contains false matches because correlation is only a heuristic and only uses local information. Inaccurate location of extracted points because of intensity variation or lack of strong texture features is another source of error. The geometric constraint between two images is the homography constraint (1). If two points are correctly matched, they must satisfy this constraint, which is unknown in this case. If the homography between the two images is estimated based on a least-squares criterion, the result could be completely wrong even if there is only one false match. This is because least-squares is not robust to outliers (erroneous data). A technique based on a robust estimation technique known as the least median squares was developed to detect both false matches and poorly located corners, and simultaneously estimate the homography matrix H. More precisely, let {(m_1i, m_2i)} be the pairs of points between two images matched by correlation, the homography matrix H is estimated by solving the following nonlinear problem:

$\begin{array}{cc}\underset{Hi}{\min \mathrm{median}}{m_{2i}-{\hat{m}}_{1i}}^2& \left(2\right)\end{array}$
where {circumflex over (m)}_1iis the point m_1itransferred to the current image by H, i.e., {circumflex over ({tilde over (m)}_1i=λ_iH{tilde over (m)}_1i.

The aforementioned optimization is performed by searching through a random sampling in the parameter space of the homography to find the parameters yielding the smallest value for the median of squared residuals computed for the entire data set. From the smallest median residual, a so-called robust standard deviation {circumflex over (σ)} can be computed, and any point match yielding a residual larger than, say, 2.5{circumflex over (σ)} is considered to be an outlier and is discarded. Consequently, it is able to detect false matches in as many as 49.9% of the whole set of matches. More concretely, after inputting a pair of images (referred to as a first image and a second image for explanation purposes) (process action802), this outlier rejection procedure is shown inFIG. 8 and is implemented as follows:

• 1. Draw m random subsamples of p=4 different point matches (process action804). (At least 4 point matches are needed to determine a homography matrix.)
• 2. For each subsample J, compute the homography matrix H_Jaccording to (1) (process action806).
• 3. For each H_J, determine the median of the squared residuals, denoted by M_J, with respect to the whole set of point matches. The squared residual for match i is given by ∥m_2i−{circumflex over (m)}_1i∥²where {circumflex over (m)}_1iis point m_1itransferred to the second image by H_J(process action808).
• 4. Retain the estimate H_Jfor which M_Jis minimal among all m M_J's (process action810).
• 5. Compute the robust standard deviation estimate: {circumflex over (σ)}=1.4826[1+5/(n−p)]√{square root over (M_J)}, where n is twice the number of matched points (process action812).
• 6. Declare a point match as a false match if its residual is larger than k{circumflex over (σ)}, where k is set to 2.5 (process action814).
• 7. Discard the false matches and re-estimate H by minimizing the sum of squared errors Σ₁∥m_2i−{circumflex over (m)}_1i∥²where the summation is over all good matches (process action816).

In one embodiment of the invention, m=70, was used, which gives a probability of 99% that one of the 70 subsamples is good (i.e., all four point matches in the subsample are good) even if half of the total point matches are bad. This last step improves the accuracy of the estimated homography matrix because it uses all good matches.

This incremental matching procedure stops when all images have been processed (process action818). Because of the incremental nature, cumulative errors are unavoidable.

2.6 Adjusting the Homographies through Global Optimization

For higher accuracy, one needs to adjust H's through global optimization by considering all the images simultaneously. This is done as follows. Let one assume that one has in total N images. Without loss of generality, the first image is chosen as the reference image for the global optimization. Let the homography matrix from the reference image to image i be H₁, with H₁=I. There are M distinct points in the reference image, which are called reference points, denoted by {circumflex over (m)}_j. Because of the matching process, a reference point is observed at least in two images. For example, a point in the first image can be matched to a point in the second image, which in turn is matched to a point in the third image; this happens if the first three images shares a common region. Even if a physical point in space is observed in three or more images, only one single reference point is used to represent it. One additional symbol φ_ijis introduced:

• • φ_ij=1 if point j is observed in image i;
    • 0 otherwise.
      One can now formulate the global optimization as estimation of both homography matrices H_i's and reference points {circumflex over (m)}_j's by minimizing the errors between the expected positions and the observed ones in the images, i.e.,

$\underset{\left\{H_i\right\},\left\{{\hat{m}}_j\right\}}{\min }\sum _{i=1}^N\sum _{j=1}^M{m_{\mathrm{ij}}-{\hat{m}}_{\mathrm{ii}}}^2,$
where {tilde under ({circumflex over (m)}_ijλ_ijH{tilde under ({circumflex over (m)}_j
with λ_ijbeing a scalar factor.
2.7 Stitching Images.

Once the geometric relationship between images (in terms of homography matrices H's) are determined, one is able to stitch all images as a single high-resolution image. There are several options, and in one working embodiment a very simple one was implemented. As shown inFIG. 9,process action902, the first image is used as the reference frame of the final high-resolution image, and original images are successively matched to the reference frame (process action904). If a pixel in the reference frame appears several times in the original images, then the one in the newest image is retained, as shown inprocess action906.

2.8 Alternate Method of Determining Matching Points of Interest.

In order to achieve higher efficiency and robustness in matching two images without knowing any information about their relative position, a pyramidal and multi-starts search strategy was developed. The pyramidal search strategy is particularly useful when the size of the input images is very large. The flow chart of this process is shown inFIG. 10.

The process works as follows. A pair of consecutive images is input, as shown inprocess action1002. A check is made as to whether the image resolution is too high (process action1004). For example, in one embodiment of the invention, if the image width or height is bigger than 500 pixels, the image resolution is considered as too high. If the resolution is too high, then the images are down sampled (process action1006). The image size is reduced by half in each iteration, and thus a pyramidal structure for each image is built, up to a level at which the image resolution reaches the desired one. At the lowest level (or with the original resolution if the size of input images is not too large), a multi-start search strategy is employed (process action1008).

The multi-start search strategy as follows. For any given pixel in one image, its maximum displacement in the other image (i.e., the maximum difference between any pair of corresponding pixels, or the maximum disparity) is the image size if one assumes there is an overlap between the two images. Considering the previously described matching and homography estimation algorithm works with relatively large unknown motion, one does not need to examine every possible displacement. Instead, the displacement (which is equal to the image size) space is coarsely sampled uniformly. More concretely, the procedure is as follows.

• • 1. Nine start points are generated, each defining the center of a search window of the previously described matching algorithm. Let W and H be the width and height of an image. The nine points are (−W/2, −H/2), (0,−H/2), (W/2, −H/2); (−W/2, 0), (0, 0), (W/2, 0); (−W/2, H/2), (0, H/2), (W/2, H/2). The size of the search window is equal to (3W/4, 3H/4), so there is an overlap between adjacent search windows in order to lower the probability of miss due to coarse sampling. Note that with this size of the search window, one does not cover the small region near the boundary, which corresponds to an overlap less than ⅛^thof the image size.
  • 2. For each start point, the aforementioned matching and homography estimation algorithm is run, which gives the number of matched points and the root of mean square errors (RMS) of matched points, as well as an estimation of the homography between two images.
  • 3. The homography estimation which corresponds to the largest number of matched points and the smallest RMS is chosen (process action1010).
    If the last level has not been reached (process action1012) one then proceeds to the higher level using the previously estimated homography, which consists of two steps:
  • 1. The homography is projected to the higher level (process action1014). Let H_i−1be the homography at level i−1. Since the images at level i is twice as big as the images at level i−1 in the pyramidal structure, the corresponding homography at level i, H_i, is equal to S H_i−1S⁻¹, where S=diag(2,2,1).
  • 2. The homography is refined at the current level (process action1016). There are at least two ways to do that.
    • Simple Technique. First, the four image corners are transformed using H_i, the disparity is computed for each point (i.e., the difference between the transformed corner point and the original one), and the maximum and minimum disparities are computed in both the horizontal and vertical directions. Second, the search range is defined by enlarging the difference between the maximum and minimum disparities by a certain amount (10% in one embodiment) to account for the imprecision of the estimation H_i. Finally, the points are matched using the search range defined earlier, and the homography is estimated based on least-median-squares.
    • Elaborate Technique. First, the first image and all the detected corners are transformed using H_i. Second, the corners are matched and the homography between the transformed image and the second image is estimated. This estimated homography is denoted by ΔH_i. The search range could be quite small (say, 15 pixels) to consider the imprecision of H_iestimated at a lower level. Finally, the refined homography is given by ΔH_iH_i.
      The above process is repeated until the original images are matched, and the estimated homography is reported as the output (process action1018).
      3.0 Exemplary Working Embodiment

The following paragraphs describe an exemplary working embodiment of the system and method of converting the content of a whiteboard, paper, or similar object into a high-resolution image.

In this section, a few examples are shown.FIGS. 11 and 12 show the stitching result of six images of a whiteboard.FIGS. 13 and 14 show the stitching result of a paper document. The stitched paper document, shown inFIG. 14A, is compared inFIG. 14B with a single image of the document, and clearly the stitched image gives a much higher readability.

Claims (17)

1. A computer-implemented process for converting the contents of a planar object into a high-resolution image, comprising the process actions of:

acquiring a sequence of images of portions of a planar object which have been captured in a prescribed pattern and wherein each subsequent image overlaps a previous image in said pattern;

extracting points of interest in each image;

matching said points of said interest between each pair of successive images wherein the process action of matching said points of said interest between each pair of successive images comprises the process actions of:

for each point of interest in the earlier captured image of the pair of images under consideration, establishing a window of pixels centered on the point of interest;

for each point of interest,

comparing said window in said earlier captured image of the pair under consideration with windows of pixels the same size which are centered on the points of interest in the later-captured image of the pair by computing a zero-mean normalized cross correlation between the windows; and

whenever the largest correlation scare computed between the window of the earlier captured image and each of the windows in the later captured image exceeds a prefixed threshold, designating the associated point in the later-captured image as a match candidate for the point under consideration in the earlier-captured image;

computing a projective mapping between each pair of successive images in order to determine corresponding pixel locations in the images; and

generating a composite image from said sequence of images using said projective mapping.

2. The computer-implemented process ofclaim 1 wherein the match candidate with the largest correlation in the earlier-captured image is considered a match when the point under consideration in the later-captured is the same as the point corresponding to the match candidate in the earlier captured image.

3. The computer-implemented process ofclaim 1 wherein said process action of computing a projective mapping between each set of two successive images in said ordered sequence of image, comprises the process actions of:

(a) inputting a first image and a second image;

(b) drawing m random subsamples of p=4 different point matches;

(c) for each subsample J, computing a homography matrix H_j;

(d) for each H_J, determining the median of the squared residuals, denoted by

M_J, with respect to the whole set of point matches, where the squared residual for match i is given by ∥m_2i−{circumflex over (m)}_1i∥²where {circumflex over (m)}_1iis point in m_1i, transferred to the second image by H_J;

(e) retaining the estimate H_Jfor which M_Jis minimal among all m M_J's;

(f) computing a robust standard deviation estimate {circumflex over (σ)};

(g) declaring a point match as a false match if its residual is larger than k{circumflex over (σ)}, where k is set to a prescribed value;

(h) discarding the false matches and re-estimating H by minimizing the sum of squared errors Σ_i1∥m_2i−{circumflex over (m)}_1i∥²where the summation is over all good matches; and

(i) repeating process actions (a) through (h) until all images of me sequence of images have been processed.

4. The computer-implemented process ofclaim 3 wherein k=2.5.

5. The computer-implemented process ofclaim 3 wherein the robust standard deviation {circumflex over (σ)}=1,4826[1+5/(n−p)]√{square root over (M_j)} where n is twice the number of the matched points.

6. The computer-implemented process ofclaim 1, wherein the process action of computing a projective mapping comprises computing a homography in the form of a homography matrix between each pair of two successive images and refining the homographies computed for each pair of successive images through global optimization; further comprising the process actions of:

for a total of N images, choosing a first image as the reference image for the global optimization and letting me homography matrix from the reference image to image i be H_i, with H₁=1, the identity matrix, and defining that there are M distinct points in the reference image, which are called reference points, denoted by {circumflex over (m)}_j; defining a variable φ_ij=1 if point j is observed in image i and φ_ij=0 otherwise;

formulating the global optimization as an estimation of both homography matrices H_i's and reference points {circumflex over (m)}_ij's by minimizing the errors between the expected positions of the reference points in the images and the observed reference points in the images.

7. The computer-implemented process ofclaim 6 where the estimation of both homography matrices H_i's and reference points {circumflex over (m)}_j's is computed by minimizing the errors between the expected positions of said reference points and the observed reference points in the images using the equation

$\underset{\left\{H_i\right\},\left\{{\hat{m}}_j\right\}}{\min }\sum _{i=1}^N\sum _{j=1}^M{m_{\mathrm{ij}}-{\hat{m}}_{\mathrm{ii}}}^2,$

where={tilde over ({circumflex over (m)}_ij=λ_ijH{tilde over ({circumflex over (m)}_jwith λ_ijbeing a scalar factor.

8. A system for converting markings on a planar object into a high resolution image, the system comprising:

a general purpose computing device; and

a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to,

acquire a sequence of images of portions of a planar object having been captured in a prescribed pattern, each subsequent image overlapping a previous image in said pattern;

extract points of interest in each image in said sequence;

match said points of said interest between two successive images in said sequence; wherein the program module for matching said points of said interest between two successive images in said sequence comprises sub-modules for,

for each point of Interest in the earlier captured image of the set of images under consideration, establishing a window of pixels centered on the point of interest,

for each point of interest,

comparing said window in said earlier captured image of the set under consideration with windows of pixels the same size which are centered on the points of interest in the later-captured image of the set by computing a zero-mean normalized cross correlation between the windows, and

whenever the largest correlation score computed between the window of the earlier captured image and each of the windows in the later captured captured image as a match candidate for the point under consideration in the earlier-captured image;

compute a projective mapping between each set of two successive images in said sequence of images in order to determine corresponding pixel locations in the images of each set; and

generate a composite image from said images using said projective mapping.

9. The system ofclaim 8 wherein the match candidate with the largest correlation in the earlier-captured image is considered a match when the point under consideration in the later-captured image is the same as the point corresponding to the match candidate in the earlier captured image.

10. The system ofclaim 8 wherein said program module for computing a projective mapping between each set of two successive images in said sequence of images, comprises sub-modules for:

(a) inputting a first image and a second image;

(b) drawing m random subsamples of p=4 different point matches;

(c) for each subsample J, computing a homography matrix H_j,

(d) for each H_J, determining the median of the squared residuals, denoted by M_J, with respect to the whole set of point matches, where the squared residual for match i is given by ∥m_2i−{circumflex over (m)}_1i∥²where {circumflex over (m)}_1iis point m_1itransferred to the second image by H_J;

(e) retaining the estimate H_Jfor which M_Jis minimal among all m M_J's;

(f) computing the robust standard deviation estimate {circumflex over (σ)}=1.4826[1+5/(n−p)]√{square root over (M_J)}where n, is twice the number of matched points;

(g) declaring a point match as a false match if its residual is larger than k{circumflex over (σ)}, where k is set to a prescribed number;

(h) discarding the false matches and re-estimating H by minimizing the sum of squared errors Σ₁∥m_2i−{circumflex over (m)}_1i∥²where me summation is over all good matches; and

(i) repeating process actions (a) through (h) until all images have been processed.

11. The system ofclaim 10 wherein the homography matrix H_Jis calculated by letting m₁=[u₁, v₁]^Tand m₂=[u₂, v₂]^Tbe a pair of corresponding points, and using the notation {tilde over ( )} for {tilde over (m)}=[u ,v1]^T, such that {tilde over (m)}₂=σH{tilde over (m)}_j.

12. A computer-readable medium having computer-executable instructions for converting a series of low resolution images of portions of a planar object into a high resolution image of said object, said computer executable instructions causing a computer to execute the method comprising:

acquiring a series of images of the depicting portions of the same scene;

extracting points of interest in each image of said series of images;

matching said points of interest in each image of said series of images with the image preceding said image in said series of images, wherein said matching said points of interest in each image of said series of images with the preceding said image in said series of images comprises,

for each point of interest in the preceding image of a pair of images under consideration, establishing a window of pixels centered on the point of interest;

for each point of interest,

comparing said window in said preceding image of the pair under consideration with windows of pixels the same size which are centered on the points of interest in the later-captured image of the pair by computing a zero-mean normalized cross correlation between the windows, and

whenever the largest correlation score computed between the window of the preceding image and each of the windows in the later captured image exceeds a prefixed threshold, designating the associated point in the later-captured image as a match candidate for the point under consideration in the preceding images;

calculating a homography between each image of said series of images with the image preceding said image in said series of images; and

stitching each image in said series of images together using said homographies to create a composite image.

13. The computer-readable medium ofclaim 12 wherein said calculating a homography between each image of said series of images with the image preceding said image in said series of images comprises solving the nonlinear problem

$\underset{Hi}{\min \mathrm{median}}{m_{2i}-{\hat{m}}_{1i}}^2$

where {(m_1i, m_2i)} are the pairs of points between two matched images, H is the homography matrix, {circumflex over (m)}_1iis the point m_1itransferred to the current image by H , and where {tilde over ({circumflex over (m)}_1i=λ_iH{tilde over (m)}_1i, λ_ibeing a scalar factor.

14. The computer-readable medium ofclaim 12 wherein the match candidate with the largest correlation in the preceding image is considered a match when the point under consideration in the later-captured image is the same as the point corresponding to the match candidate in the preceding image.

15. The computer-readable medium ofclaim 12 wherein said matching said points of interest in each image of said series of images with the preceding image in said series of images further comprises:

(a) inputting a pair of consecutive images of a set of original images;

(b) checking whether the image resolution is greater than a prescribed value;

(c) if the resolution is too high, then downsampling the images by half at each iteration that the resolution is too high, thus building a pyramidal structure for each image up to a level at which the image resolution reaches a prescribed resolution;

(d) at the lowest level, or with the original resolution if the size of input images is not too large, employing a multi-start search strategy is employed to estimate a homography between the pair of images;

(e) generating nine start points, each defining the center of a search window of the previously described matching algorithm, where W and H are defined as the width and height of an image and the nine points are (−W/2, −H/2), (0, −H/2), (W/2, −H/2); (−W/2, 0), (0, 0), (W/2, 0); (−W/2, H/2), (0, H/2), (W/2, H/2);

(f) for each start point, running a matching and homography estimation procedure, which gives the number of matched points and the root of mean square errors (RMS) of matched points, as well as an estimation of the homography between the two images;

(g) choosing the homography estimation which corresponds to the largest number of matched points and the smallest RMS;

(h) if the last level has not been reached, proceeding to the higher level using the previously estimated homography;

projecting the homography to the higher level, letting H_i-1be the homography at level i−1 and the corresponding homography at level i, setting H_i, equal to S H_i−1S⁻¹, where S=diag(2, 2, 1); refining the homography at the current level; and

(i) repeating (a) through (h) until all images of the set of original images are matched, and reporting the last estimated homography as the output.

16. The computer-readable medium ofclaim 15 wherein said refining said homography comprises

transforming the four image corners using H_j;

computing the disparity for each point wherein the disparity is the difference between the transformed corner point and the original one;

computing the maximum and minimum disparities in born the horizontal and vertical directions;

defining the search range by enlarging the difference between the maximum and minimum disparities by a certain amount to account for the imprecision of the estimation H_j;

matching the points using said defined search range; and

estimating the homography based on least-median-squares.

17. The computer-readable medium ofclaim 15 wherein said refining said homography comprises:

transforming the first image and all the detected corners using H_j;

matching the corners and estimating the homography between the transformed image and the second image, said estimated imaged being denoted by ΔH_j, calculating the refined homography as ΔH_jH_i.

Images